An ANOVA is being used to test differences between four groups. The main alternative and null hypotheses for this test can be written as follows: HA:MA FMB *HC*MD. Ho: HA = MB = H0 = HD If you wanted to do post-hoc (aka follow-up) tests to figure out exactly which groups statistically differed from each other, how many would you need to do